This proposal was prepared in response to NIH Program Announcement (PA-98-031) entitled "Methodology and measurement in the behavioral and social sciences." Research on autobiographical memory has shown that numerous cognitive processes have the potential to bias retrospective reports of personal experiences. Behavioral and medical scientists and practitioners rely heavily on recall procedures in their research endeavors and for treating patients. Measures of symptoms, pain, treatment efficacy, adherence, and traumatic experiences, for example, are typically based on retrospective data, sometimes for periods of several months or even years. The primary aims of this proposal involve comparing assessments of momentary experiences, based on a procedure we have called Ecological Momentary Assessment (EMA), with those of recalled experiences in an effort to determine 1) the relationship between the two types of data and 2) the meaning and practical utility of each type of data. The first aim seeks to understand how people summarize their experiences and create a recall. It will also explore situational influences and characteristics of the person that influence recall ratings. The second aim is grounded in the hypothesis that summaries of experiences based upon momentary ratings versus recall carry information that may be useful for different purposes, which could have important theoretical and practical consequences for scientists and practitioners. Secondary aims investigate several methodological issues pertain to the conduct of EMA studies including potential reactivity effects, participant burden, frequency of momentary sampling, optimal statistical methods for analyzing EMA data, and methods for improving recall of experiences. Outcome variables for testing these questions are mood and pain. Both are constructs with wide basic and applied utility in behavioral and medical domains. These measures will be intensively examined in patients with osteoarthritis and rheumatoid arthritis, some with recent diagnoses and others with chronic conditions in order to ensure variability in mood and pain over time. Results from these studies should have implications for understanding the meaning of recalled mood and pain measures, addressing the questions: how do people summarize their experiences and what factors influence those summaries? They will also increase knowledge about when momentary versus recalled summary measures should be used for addressing basic and applied questions. Finally, the study will provide practical information for researchers using momentary assessment methods.